Portfolio and Diversification

TOPICS

1. Expected return of portfolio and Risk

2. Diversification effects of portfolio
A Little History of Portfolio Theory

▪ In March 1952, Harry Markowitz, a 25 year old graduate

student from the University of Chicago, published ”Portfolio
Selection” in the Journal of Finance.
▪ The paper opens with: ”The process of selecting a portfolio
may be divided into two stages. The first stage starts with
observation and experience and ends with beliefs about the
future performances of available securities. The second
stage starts with the relevant beliefs about future
performances and ends with the choice of portfolio”.
▪ Thirty eight years later, this paper would earn him a Nobel
Prize in economic science.
1. Expected Return and Risk of Portfolio
▪ Portfolio:
– Sets of invested assets for a certain investor
▪ Why we concern portfolio ?
– Reduce risk and increase expected return
simultaneously using Diversification.
▪ Portfolio formation processes
1. Calculate expected return and risk of portfolio
▪ Expected return of each asset, volatility, and covariance
between assets.
2. Reduce the choice set of portfolio based on
dominance principle
3. Determine optimal portfolio using risk attitude of a
investor
Expected Return and Risk of Portfolio

▪ Covariance of Returns
– A measure of the degree to which two variables
move together relative to their individual mean
values over time.

– Cov ij = E{[R i − E(R i )][R j − E(R j )]}

n

 (R it − Ri )( Rtj − R j )
= t =1
N −1
Expected Return and Risk of Portfolio
▪ Calculation of expected return
Expected Return and Risk of Portfolio

▪ Computation of covariance
Expected Return and Risk of Portfolio

▪ Covariance vs. correlation

– Limitations of covariance
▪ Know direction of linear co-movement but hard to
know the degree of co-variability
– Correlation
▪ Standardized covariance measure

C ov ij
▪ ρ ij =
σ iσ j
Expected Return and Risk of Portfolio

▪ Computation of Correlation

C o v ij 6 .3 7
ρ ij = = = 0 .1 0 8
σ iσ j ( 5 . 8 0 )(1 0 . 1 7 )
Two-Risky-Assets Portfolio

▪ Expected return
E ( r p ) = w B E ( rB ) + w S E ( rS )
where w1+w2 =1
▪ Expected variance

v a r( rp ) = ( w B B )2 + (w S S )2 + 2 (w B B )( w S S ) B S

▪ Minimum variance portfolio

 S2 −  B  S  B S
wB = 2
 S +  B2 − 2  B  S  B S
Two-Risky-Assets Portfolio

▪ Example
E ( rB ) = 5 %  B = 8%
E ( rS ) = 1 0 %  S = 1 9 %  B S = 0 .2

– What is the expected return and variance, if Ws is 40% and WB is

60%?
– What is the minimum variance portfolio?
– What is the expected return of portfolio, if you want your std of
portfolio of 10%?
– What is the expected variance, if you want your expected return of
portfolio of 9%?
3 Risky Assets Portfolio

▪ Expected return
E ( r p ) = w1 E ( r1 ) + w 2 E ( r2 ) + w 3 E ( r3 )
where w1+w2+w3 =1

▪ Assets portfolio’s covariance is

cov( ra , rb ) = w1a w1b 1,1 + w1a w 2b 1, 2 + w1a w 3b 1, 3 +
w 2a w1b 2 ,1 + w 2a w 2b 2 , 2 + w 2a w 3b 2 , 3 +
w 3a w1b 3 ,1 + w 3a w 2b 3 , 2 + w 3a w 3b 3 , 3
n Risky Assets Portfolio

▪ Expected return
E ( rp ) = w1 E ( r1 ) + w2 E ( r2 ) + .....wn E ( rn )
n n
=  wi E ( ri ) where w i =1
i =1 i =1

▪ Expected variance
 N  N N N
 2
p = var   w i E ( ri )  = w i
2
 i
2
+ ww i j cov( ri , r j )
 i =1  i =1 i =1 j =1
ji
N N N N
= ww
i =1 j =1
i j cov( ri , r j ) = ww
i =1 j =1
i j i, j

where cov( ra , rb ) = cov( w1a r1 +    + wNa rN , w1b r1 +    + wNb rN )

= w1a w1b 1,1 + w1a w2b 1, 2 +    + wNa wNb  N , N
N N
= w
i =1 j =1
a
i w bj i , j
2. Diversification Effects of Portfolio

▪ 2 Risky Assets
E ( rB ) = 5 %  B = 8%
E ( rS ) = 1 0 %  S = 1 9 %  B S = 0 .2
– Std (rp) with 50% for assets B and 50% for S:
11.03% Return
12

– Less than 13.5% 10

0
0 5 10 15 20
Risk
Efficient Frontier in Two Risky Assets

weight(A) weight(B) E(rP) Std(rP)

-0.5000 1.5000 0.0250 0.1373
-0.4000 1.4000 0.0300 0.1221
-0.3000 1.3000 0.0350 0.1081 E(rs) E(rB) std (rS) Std(rB) Corr(B,S)
Return
-0.2000 1.2000 0.0400 0.0959 0.1 0.05 0.19 0.08 0.2
-0.1000 1.1000 0.0450 0.0862
0.14
0.0000 1.0000 0.0500 0.0800
0.0932 0.9068 0.0547 0.0780
0.12
0.1000 0.9000 0.0550 0.0781
0.2000 0.8000 0.0600 0.0807
0.1
0.3000 0.7000 0.0650 0.0875
0.4000 0.6000 0.0700 0.0977
0.08
0.5000 0.5000 0.0750 0.1102
0.6000 0.4000 0.0800 0.1244
0.06
0.7000 0.3000 0.0850 0.1398
0.8000 0.2000 0.0900 0.1560
0.04
0.9000 0.1000 0.0950 0.1728
1.0000
1.1000
0.0000
-0.1000
0.1000
0.1050
0.1900
0.2075
0.02
1.2000
1.3000
-0.2000
-0.3000
0.1100
0.1150
0.2253
0.2433
0
1.4000 -0.4000 0.1200 0.2615 0 0.1 0.2 STD 0.3
1.5000 -0.5000 0.1250 0.2798
Effects of correlation in Diversification

STD
weight(A) weight(B) E(rp)
roh=1 roh=0.5 roh=0 roh=-0.5 roh=-1
-0.5000 1.5000 0.0250 0.0250 0.1097 0.1531 0.1866 0.2150
-0.4000 1.4000 0.0300 0.0360 0.0990 0.1354 0.1638 0.1880
-0.3000 1.3000 0.0350 0.0470 0.0902 0.1186 0.1414 0.1610
-0.2000 1.2000 0.0400 0.0580 0.0837 0.1032 0.1196 0.1340
-0.1000 1.1000 0.0450 0.0690 0.0802 0.0900 0.0989 0.1070
0.0000 1.0000 0.0500 0.0800 0.0800 0.0800 0.0800 0.0800
0.0932 0.9068 0.0547 0.0903 0.0828 0.0747 0.0655 0.0548
0.1000 0.9000 0.0550 0.0910 0.0831 0.0745 0.0646 0.0530
0.2000 0.8000 0.0600 0.1020 0.0893 0.0744 0.0557 0.0260
0.3000 0.7000 0.0650 0.1130 0.0979 0.0799 0.0565 0.0010
0.4000 0.6000 0.0700 0.1240 0.1083 0.0899 0.0666 0.0280
0.5000 0.5000 0.0750 0.1350 0.1201 0.1031 0.0826 0.0550
0.6000 0.4000 0.0800 0.1460 0.1329 0.1184 0.1018 0.0820
0.7000 0.3000 0.0850 0.1570 0.1465 0.1351 0.1228 0.1090
0.8000 0.2000 0.0900 0.1680 0.1606 0.1528 0.1447 0.1360
0.9000 0.1000 0.0950 0.1790 0.1751 0.1712 0.1671 0.1630
1.0000 0.0000 0.1000 0.1900 0.1900 0.1900 0.1900 0.1900
1.1000 -0.1000 0.1050 0.2010 0.2051 0.2092 0.2131 0.2170
1.2000 -0.2000 0.1100 0.2120 0.2204 0.2286 0.2364 0.2440
1.3000 -0.3000 0.1150 0.2230 0.2359 0.2482 0.2598 0.2710
1.4000 -0.4000 0.1200 0.2340 0.2515 0.2679 0.2834 0.2980
1.5000 -0.5000 0.1250 0.2450 0.2673 0.2878 0.3070 0.3250
Effects of correlation in Diversification in 2 Risky Assets

▪ If correlation 1
var( r p ) = ( w B  B ) 2 + ( wS S ) 2 + 2 ( w B B )( w S  S )  BS
= ( w B B ) 2 + ( wS S ) 2 + 2 ( w B B )( w S  S )
= ( w B B + wS S )2

std (r p ) = ( w B  B + w S  S )
0.14
return
0.12

0.1

0.08

0.06

0.04

0.02

0
0 0.05 0.1 0.15 0.2 0.25 0.3

Standard Deviation
Effects of correlation in Diversification in 2 Risky
Assets
▪ If correlation 0
va r( r p ) = ( w B  B ) 2 + (w S S ) 2 + 2 ( w B B )( w S  S )  BS
= ( w B B ) 2
+ (w S S )2
std (rp ) = (w B B ) 2
+ (w S S ) 2

0.14
return
0.12

0.1

0.08

0.06

0.04

0.02

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Standard Deviation
Effects of correlation in Diversification in 2 Risky
Assets

▪ If correlation -1
var( r p ) = ( w B  B ) 2 + ( w S  S ) 2 + 2 ( w B  B )( w S  S )  B S
= ( w B  B ) 2 + ( w S  S ) 2 − 2 ( w B  B )( w S  S )
= ( w B B − w S  S ) 2
std (rp ) = w B B − w S S
0.14

return 0.12

0.1

0.08

0.06

0.04

0.02

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Standard Deviation
Effects of correlation in Diversification in 2 Risky
Assets
❑ Diversification
❑ Although the
expected rate of
return of any portfolio
is simply the weighted
average returns, but it
is not true of the std.
❑ Potential benefits
from diversification
arise when
correlation is less
than perfectly
positive.
Diversification in many risky assets

▪ Let’s add a fourth security to this example.

Security D has E(rD) = 15%, Std(rD)= 45%.
Assume it is a zero correlation with all of the
other securities (A, B and C)
▪ Will
Asset E(r)
anyone Std Assets
hold this A
security? B C
A 0.05 0.10 A 1.0 0.0 0.5
B 0.10 0.20 B 1.0 0.5
C 0.15 0.30 C 1.0
Diversification in Many Risky Assets
Power of Diversification

▪ Start with our equation for variance:

N N N
 2
p = w
i =1
i
2
 +
i
2
ww
i =1 j =1
i j c o v( ri , r j )
ji

▪ Then make the simplifying assumption that wi = 1/N for all

assets:
 1 N 2 N  1 N
 =  2    i +   2   cov( ri , r j )
2
p
 N  i =1 i =1  N  j =1
j i
▪ Next, because the average variance and covariance of the
securities are:
 1 N 2 1 N
 =   i , cov = 
2
cov( ri , r j )
 N  i =1 N ( N − 1) j =1
j i
Power of Diversification
▪ Plugging these equations into previous equation gives:
 1  N −1
 2
p = 
2
+ cov
N  N
▪ What happens as N becomes large?
 1  N −1
 → 0 a nd → 1
 N  N
▪ Only the average covariance matters for large portfolios.
✓ If the average covariance is zero, then the portfolio variance is close to
zero for large portfolios
Power of Diversification

Number of Number of Number of

Stocks Stocks Stocks

1 46.619 20 9.036 100 7.453

2 26.839 25 8.640 200 7.255
4 16.948 30 8.376 300 7.190
6 13.615 35 8.188 400 7.157
8 12.003 40 8.047 500 7.137
10 11.013 45 7.937 1000 7.097
14 9.883 50 7.849
16 9.530 75 7.585
Systematic Risk and Unsystematic Risk

 Diversifiable or non-
systematic risk:
✓ The component of risk
that can be diversified
away with portfolio
investment.
✓ Firm specific risk
 The Systematic risk.
✓ The risk that cannot
be diversified away
the systematic risk.
✓ Related with general
market condition
Questions ?